29 July 2021
AperTO - Archivio Istituzionale Open Access dell'Università di Torino
Original Citation:
Progressive fragmentation of a traditional Mediterranean landscape by hazelnut plantations: The impact of CAP over time in the Langhe region (NW Italy)
Published version:
DOI:10.1016/j.landusepol.2013.08.018
Terms of use:
Open Access
(Article begins on next page)
Anyone can freely access the full text of works made available as "Open Access". Works made available under a Creative Commons license can be used according to the terms and conditions of said license. Use of all other works requires consent of the right holder (author or publisher) if not exempted from copyright protection by the applicable law.
Availability:
This is the author's manuscript
iris-AperTO
University of Turin’s Institutional Research Information System and Open Access Institutional Repository
This is the author's final version of the contribution published as:
[Godone D., Garbarino M., Sibona E., Garnero G., Godone F. 2014. Progressive
fragmentation of a traditional Mediterranean landscape by hazelnut plantations:
The impact of CAP over time in the Langhe region (NW Italy). Land Use Policy 36:
259‐266. ISSN: 02648377, doi: 10.1016/j.landusepol.2013.08.018.]
The publisher's version is available at:
[http://www.sciencedirect.com/science/article/pii/S0264837713001683]
When citing, please refer to the published version.
Link to this full text:
[http://hdl.handle.net/2318/138282
This full text was downloaded from iris‐Aperto: https://iris.unito.it/1
Progressive fragmentation of a traditional Mediterranean landscape by hazelnut plantations:
1
the impact of CAP along time in the Langhe region (NW Italy)
2 3
Abstract
4
Land use change is strongly modifying the traditional landscape of hilly productive Mediterranean 5
sites. An example of this situation is the Langhe region (Piemonte, NW Italy), where woody 6
plantations such as vineyards and orchards have been cultivated on hillslopes for centuries. In this 7
paper we assess the landscape changes occurred in the Diano study area (2651 ha) in the 1954-2000 8
period and we ascertain land use transition paths and rates in this rural ecosystem. Land use 9
mapping obtained from object-oriented analysis of aerial photographs was used to quantify land use 10
changes between 1954 and 2000. To examine the spatio-temporal patterns of land use change over 11
time, a set of spatial statistics that capture different dimensions of landscape change was identified. 12
An increase of landscape heterogeneity from 1954 to the present was observed due to the expansion 13
of orchards and the fragmentation of field crops. A significant portion (55%) of current orchards 14
surface is represented by former field crops, 24% by vineyards and 15% by forests. The strong 15
expansion of hazelnut orchards concurred to the fragmentation of the traditional rural landscape 16
dominated by vineyards, field crops and forests. Hazelnut orchards expansion was mainly located in 17
places where the cultivation of grapes was less remunerative. A further expansion of hazelnut in the 18
area should be planned, discussed and carefully monitored through change detection studies in order 19
to avoid potential unsustainable use of the land. 20
Key words
21
Historical aerial photograms; land use change; Corylus avellana L.; landscape metrics; spatial 22 pattern 23 24 1. Introduction 25
2
Rural landscapes in Mediterranean Europe have been managed and modified by people for 26
centuries. These human dominated landscapes experienced both intensification and extensification 27
of agricultural practices that are strong drivers of land use and land cover changes (Turner and 28
Gardner, 1990). In most productive sites agricultural activities (e.g. intensive cultivation and forest 29
logging) and urban sprawl increased (Stoate et al., 2001). In more marginal areas traditional rural 30
practices declined or ceased causing the abandonment of agricultural lands (Bonet, 2004; Sluiter 31
and de Jong, 2007) and favouring a subsequent reforestation process (Sitzia et al., 2010). Rural 32
European areas have recently undergone several important socio-economic changes that influenced 33
their landscape dynamics. From the beginning of the twentieth century urban areas development 34
increased to the detriment of rural ones (Antrop, 2004) causing a progressive abandonment of such 35
regions. In addition to the rural-urban migration phenomenon, between the second half of the 36
nineteenth century and the first decades of the following one, about 40 millions of European 37
workers moved to more industrialized countries (Hatton and Williamson, 1994). In the second half 38
of the twentieth century, as a consequence of post-war dynamics, a local growth of industrial 39
districts was observed (Fauri, 1996; Becattini and Coltorti, 2006). An opposite process was recently 40
observed in Europe, where a decrease in industrial employees was caused by global economical 41
changes such as the outstanding development of late-industrializing countries (Amsden, 1991) and 42
the restoration of some abandoned marginal rural areas (Pinto-Correia, 2000). The restoration of 43
rural areas was also favoured and often sustained by institutional financial support (Meeus et al., 44
1990; Vos and Meekes, 1999). 45
Land use change drivers such as urbanization, globalization and population growth are translated by 46
policy-makers into land use regulations at regional, national or supra-national scale (Van Rompaey 47
et al., 2001), but environmental conditions such as vegetation, soil, topography and climate act as 48
local constraints for the landscape pattern. The establishment in Europe of a common agricultural 49
policy (EU-CAP) is thus considered a fundamental factor influencing rural landscape change 50
(Rabbinge and Van Latesteijn, 1992; Tanrivermis, 2003). Up to 1992 EU-CAP was a production-51
3
oriented subsidy policy aimed to guarantee self-sufficiency in basic foodstuffs (Martinez-52
Casasnovas et al., 2010). Through the 1992 EU-CAP reform EU supported the farmers relative to 53
set-aside land on their farm (Van Rompaey et al., 2001). The EU-CAP reform in 2003 introduced 54
an agricultural policy that supported the long-term livelihood of rural areas focusing the attention on 55
conservation agriculture and sustainable farming (European Commission, 2004; Martinez-56
Casasnovas et al., 2010). The effects of the EU-CAP subsidies in Europe have been diverse 57
depending on local rural policies and due to the varied environmental conditions of the rural 58
landscape. 59
There are several methods and tools for land use change assessment or change detection. Among 60
them the analysis of historical geographical sources is a preferred tool to reconstruct traditional land 61
mosaic (McClure and Griffiths, 2002, Cullotta and Barbera, 2011). The most common historical 62
geographical data sources are aerial images (Schiefer and Gilbert, 2007) and cadastral maps 63
(Vuorela et al., 2002), but also old satellite images acquired for military purposes are used 64
(Scardozzi, 2008). These geographical data sources can be employed separately or combined, 65
according to their availability for a certain study area. In change detection studies the adoption of 66
aerial images is preferred as they allow a direct interpretation of landscape elements, rather than 67
historical maps where land is represented by symbols (Morgan et al., 2010). 68
In this paper we study land use change in a hilly region of southern Piemonte (north-western Italy). 69
The choice of our study area is motivated by the following considerations: (i) the landscape is a 70
traditional complex mosaic of forests, vineyards, orchards and other cultivated fields; (ii) the land 71
use changes in the area are particularly strong; (iii) we had the opportunity to use the camera 72
calibration certificates to perform a rigorous photogrammetric processing of the historical images; 73
(iv) the entire Langhe region is famous for the quality of its wine and food and is candidate to be 74
included in the UNESCO World Heritage List. 75
The goals of our study are: (i) to quantify and analyze the landscape changes occurred in the area 76
between 1954 and 2000; (ii) to determine the location of land cover change and the type of change; 77
4
(iii) to determine transition paths and transition rates of the land use categories in this ecosystem. 78
Finally, the potential effects of European and local agricultural policies on the rural landscape of 79
Langhe region are discussed in relation to the sustainability of the modern farming systems.
80 81
2. Materials and methods
82 83
2.1. Study area
84
Land cover changes and landscape structure has been studied in a 2651 ha area (Fig. 1) located in 85
the southern part of Piemonte region, north-western Italy (44°40' N, 07°59' E). The elevation ranges 86
from 190 to 634 m a.s.l. and the climate belongs to type Cfa in areas lower than 500 m a.s.l., having 87
humid summer and dry winter seasons and to type Cfb with milder conditions at upper elevations, 88
in terms of Köppen-Geiger’s classification (Peel et al., 2007). Annual precipitation ranges from 800 89
to 1100 mm with a main minimum in July and a secondary one in winter and with a peak in 90
autumn. Total annual rainfall averages 730.4 mm and mean annual temperature averages 11.9 °C 91
(Rodello climatic station, 415 m a.s.l.). Lithological substrate is made up of siltstone and marlstone 92
on hillsides and alluvial deposits in valley bottoms and soils are mainly represented by Entisols and 93
Inceptisols (ARPA, 2012). 94
The study area (Diano) is part of Langhe hilly region which is characterized by strong agricultural 95
character and is widely renowned for its high quality wine production like Barolo and Barbaresco 96
(Delmastro, 2005). The entire Langhe region is currently candidate to be included within the 97
UNESCO World Heritage List (UNESCO, 2012). The study site falls within the municipalities of 98
Diano d’Alba (52%), Alba, Grinzane Cavour, Serralunga d'Alba, Rodello, Benevello and 99
Montelupo Albese. We used the demographic data of Diano d’Alba and the agricultural statistics of 100
the entire Cuneo Province in order to describe the demographic trends experienced by the analyzed 101
land surface. The population density declined in the first part of the studied period (1951-1971) 102
5
from 2612 to 2216 respectively, but consistently increased in the last decades from 2216 in 1971 to 103 2980 in 2001. 104 105 2.2. Image analysis 106
Historical aerial photographs were retrieved in the archive of Italian National Research Council 107
(CNR-IRPI, Torino), where historical and recent aerial images concerning hydrogeological 108
phenomena are stored (IRPI, 2012). During the archive consultation a small block of four 109
photograms has been recovered. These aerial photographs belong to the Gruppo Aeronautico 110
Italiano (GAI) flight that represents the first available flight covering almost all Italian territory after 111
the Second World War. Older aerial images for Italian territory are available from the beginning of 112
the twentieth century. Both USAAF (Today USAF) and RAF employed images for 113
photointerpretation purposes in order to plan bombing or army raids (Kaye, 1957). These images 114
were also processed and merged with the aim of making three-dimensional terrain models of the 115
theatre of operations (Reed, 1946). On the other side Luftwaffe and Regia Aeronautica performed 116
several flights for intelligence purposes and to monitor Allied campaign in southern Italy (Ceraudo, 117
2005). GAI flight was carried out in 1954-55 with a flight height ranging from 10000 m a.s.l. in 118
mountain regions to 5000 m a.s.l. in the plains, and having a medium scale of 1:33000 (Campana 119
and Francovich, 2003; Acosta et al., 2005; Beni Culturali, 2011). 120
We had the extraordinary opportunity to find the camera calibration certificates of the investigated 121
flight at the Italian Military Geographic Institute (IGMI) historical archive allowing us to perform a 122
rigorous image orientation. Each photogram was scanned in TIF format to 600 dpi resolution, and 123
its orientation was obtained through the Automatic Aerial Triangulation approach (Mikhail et al., 124
2001) and the employment of the above-mentioned certificates assuring an overall accuracy of 2.22 125
meters. The oriented images were then orthorectified and mosaicked at 1-m resolution. Because the 126
calibration certificates for GAI flight are usually rare, we assessed their role in the process by 127
computing a second orientation employing only the focal length value. Through a comparison of the 128
6
obtained residuals we observed a quality loss of an order of magnitude when the process is carried 129
out without calibration data (Table 1). The whole image processing was accomplished using Z-Map 130
software. A recent, RGB, orthoimage (Terraitaly - IT2000™, Blom C.G.R. S.p.A) having a nominal 131
scale of 1:10000 and a ground resolution of 1-m, was employed in the change detection analysis. 132
133
2.3. Image classification
134
Automated segmentation with eCognition (scale parameter = 100, shape factor = 0.5) with manual 135
correction was used to delineate polygons on the test area. The segmented images were on-screen 136
classified into six categories of land cover (Table 2). The two resulting maps (1954 and 2000) were 137
then enhanced in a GIS environment in order to reduce the effect of different input image quality 138
and achieve a minimum mapping unit (MMU) of 100 m2. At the end of the above-mentioned phase
139
an additional topological check was performed by merging adjacent polygons of the same land 140
cover category (Fig. 2). 141
142
2.4. Landscape analysis
143
Landscape changes in the studied period (1954-2000) were assessed through change detection 144
approach and a comparison of landscape metrics over time. The change detection analysis on the 145
two land cover maps was performed by using the “Change detection” free extension in ArcView 146
environment (Chandrasekhar, 1999). This GIS extension allowed computing a transition matrix 147
reporting the transition between each pair of land cover categories as extent or proportion of area 148
per unit time. 149
To analyze changes in landscape pattern, we used Fragstats software (McGarigal and Marks, 1995) 150
to calculate several key landscape metrics for the studied period, applying an 8-cell neighbourhood 151
definition. We selected representative metrics for landscape configuration and composition, 152
including patch size and density, edge, contagion, connectivity, and diversity (Cushman et al., 153
2008). Since many metrics are closely related at the landscape level and describe similar aspects of 154
7
landscape structure (Riitters et al., 1995; Cain et al., 1997; Neel et al., 2004), ten landscape-level 155
metrics were selected excluding those that were highly correlated (r > 0.8) (Tischendorf, 2001). 156
Landscape structure was also analyzed at the class level by computing 13 metrics for the 157
6 land cover classes, for the two time periods. 158 159 3. Results 160 161 3.1. Landscape structure 162
An accuracy assessment was performed on each land use map resulting in the K statistic (Landis 163
and Koch, 1977) ranging from 0.86 (90.2% overall accuracy) for the 1954 image to 0.87 (90% 164
overall accuracy) for the 2000 image. Our analyses on landscape structure showed that important 165
changes have been occurred at Diano study site during the 1954-2000 period. A general increase in 166
landscape heterogeneity from 1954 to the present was observed. The metrics computed for the 167
landscape as a whole (Table 3) showed an increase in patch density (PD) and a decrease of patch 168
area (AREA_MN, LPI). A reduction of shape complexity (Shape Index, Contiguity Index) was 169
confirmed by a reduction of Edge Density (ED). Patches aggregation (CONTAG) decreased and a 170
decline in the isolation of patches of the same category (ENN_MN) was also observed. Patch 171
richness (six categories) remained unchanged during the observed period, but diversity (Simpson’s 172
Diversity Index) slightly increased. 173
174
3.2. Landscape change
175
The change detection approach highlighted remarkable changes in study area land use. ‘Fields’ and 176
‘Orchards’ land cover categories experienced the strongest variations (Fig. 3). The total surface of 177
‘Orchards’ increased of 24.6%, instead the ‘Fields’ category strongly (-26.9%) decreased. A slighter 178
increasing tendency (3.6%) was observed for the ‘Urban’ class too. A relatively little change was 179
observed for both the ‘Forests’ and ‘Vineyard’ categories that experienced an increase and a 180
8
reduction respectively. Based on transitions occurring in the period from 1954 to 2000 (Table 4), 181
five main transformations can be highlighted: 182
Fields Orchards (375 ha), Fields Vineyards (269 ha), Vineyards Orchards (165 ha), Fields 183
Forests (161 ha), Forests Orchards (105 ha). 184
The establishment of new settlements took place at the expenses of fields, forests and vineyards 185
mainly. The noteworthy variation experienced by the ‘Orchards’ category pushed us to deepen our 186
analysis on class level transitions. Only 3% of the total surface of ‘Orchards’ category remained 187
unchanged, while 55% of them were former fields, 24% vineyards, 15% forests and 3% other 188 categories. 189 190 4. Discussion 191 192 4.1 Landscape changes 193
Land use change study requires careful approaches able to deal with the heterogeneity of the 194
involved tools (e.g. resolution) and the variability of the landscape processes (e.g. spatial and 195
temporal scale). Such an investigation should be carried out by employing trustworthy data sources 196
in order to correctly reconstruct historical landscape patterns (Burgi and Russell, 2001). For this 197
reason, we adopted a rigorous photogrammetric approach that involved camera calibration 198
certificate assuring accurate orthorectification results (Rocchini et al., 2012). In particular was 199
possible to obtain an orientation quality for the GAI flight images higher than those from previous 200
studies (Peroni et al., 2000; Gennaretti et al., 2011). The adopted image processing approach 201
together with the obtained high classification accuracy assured a reliable land use change analysis. 202
Among the six land use categories, ‘Orchards’ increased from 1954 to 2000, replacing mainly other 203
agricultural areas (‘Fields’ and ‘Vineyards’) and ‘Forests’. During the same period, ‘Fields’ 204
drastically decreased and were replaced mainly by ‘Orchards’, ‘Vineyards’ and ‘Forests’. The 205
strong expansion of ‘Orchards’, together with the increase of ‘Urban’ areas (Chiabrando et al., 206
9
2009, 2011) transformed a traditional landscape that was dominated by vineyards, crops and forests 207
in a more fragmented land mosaic represented by a higher evenness between the land use 208
categories. Moreover, a decrease of patch shape complexity was observed at landscape level 209
(Contiguity and Landscape Shape indices) and this was probably due to the regular boundaries of 210
the new hazelnut plantations. Particularly interesting is the overall dynamic of forests and vineyards 211
that remained almost constant in terms of total surface but experienced a substantial change. In the 212
case of forests, the natural reforestation process confined to the more marginal areas contrasted the 213
expansion of orchards. The reforestation pattern observed at Diano study sites is in agreement with 214
other Italian and European (Falcucci et al., 2007; Sitzia et al., 2010; Cocca et al. 2012) mountainous 215
and hilly areas, but the productive nature of the site greatly limited the trees encroachment. The 216
nonmarginal character of the investigated area is also witnessed by a remarkable increase of 217
inhabitants observed in the last decades. This trend is in contrast with many other sites in Italy and 218
other EU countries (Pinto-Correia, 1993; Peroni et al., 2000; Conti and Fagarazzi, 2004; Zomeni et 219
al., 2008). As opposite to forests, the vineyards expanded in the most accessible and productive 220
sites, thus limiting the orchards expansion too. However a reduction of vineyards surfaces in 221
marginal and less accessible sites in favour of orchards was observed and confirmed by other 222
studies in the Mediterranean region (e.g. Marull et al., 2010; Corti et al., 2011). On the contrary, 223
there are other Mediterranean-climate ecoregions where a strong expansion of vineyards was 224
favoured by wine market booming (Merenlender, 2000; Fairbanks et al. 2004) or agricultural 225
policies (Cots-Folch et al., 2006). 226
227
4.2 Transition from vineyards to hazelnut orchards
228
The land use category defined as ‘Orchards’ in the present paper was almost entirely represented by 229
hazelnut (Corylus avellana L.) orchards. A wider classification that included all the orchards was 230
used in this paper in order to reduce misclassification errors (Franco, 1997). The domestication of 231
hazelnut in Mediterranean areas probably started during the Greeks and Romans periods (Trotter, 232
10
1921; Boccacci and Botta, 2009), but became important for the Langhe region at the end of 1800 233
(Comunità Montana Alta Langa, 2009). At that time the appearance of downy mildew (Plasmopara 234
viticola [Berk. & Curt] Berl. & de Toni), of grape phylloxera (Daktulosphaira vitifoliae Fitch) and
235
other grapevine parasites increased the uncertainty of winemakers that started to cultivate the 236
hazelnut (Valentini and Me, 2002). A similar dynamic has been observed also in the metropolitan 237
region of Barcelona, Spain (Marull et al., 2010). During the Second World War the hazelnut oil was 238
used as a surrogate for olive oil, but only in the late 80s its cultivation became really important and 239
expanded in Piedmont region. The hazelnut cultivated surface increased by 20% in the last decade 240
of the study period, and the highest increment peak was observed during the 1990 – 1995 period 241
(Valentini and Me, 2002). Moreover, during the 1981 – 2000 period the surface expansion triggered 242
an increase of hazelnut production from 9.171 tons to 11.959 tons and of its price from 1.66 €/Kg to 243
1.96 €/Kg. A shift of vineyard cultivation toward hazelnut is detectable from the historical statistics 244
on cultivated surfaces of the Cuneo province (ISTAT 1971-2001). In a twenty year time span (1980 245
- 2000) hazelnut orchards have nearly doubled their surfaces, while vineyards have shown a 246
remarkable decrease (Fig. 4). These historical statistics confirmed the results observed at Diano site 247
through change detection analysis. 248
249
4.3 Agricultural policies concerning hazelnut cultivation
250
The strong increase of hazelnut surfaces occurred in Piedmont at the beginning of the 90s, reflected 251
an increasing interest of landowners towards hazelnut and its economical potential (Valentini and 252
Me, 2002). Even if we did not directly measure the effects of local and European agricultural 253
policies on the hazelnut cultivation, it is interesting to mention those events that influenced its 254
diffusion in Piedmont region. The European Union supported young farmers through a regulation 255
devoted to improve the efficiency of agricultural structures (EC 2328/91) and encouraged the 256
adoption of agronomic practices with a positive impact on the environment, through the agri-257
environment regulation (EU 2078/92). Probably the most important policy measure regarding 258
11
hazelnut in Piedmont is a decree of the Italian Ministry of Agricultural and Forestry Policies (DM n. 259
2/12/93) that recognized its Protected Geographical Indication (PGI) under the appellation 260
“Nocciola Piemonte”. In 1996 the European Union registered the “Piedmont hazelnut” as a 261
Protected Geographical Indication (PGI) through a regulation on the registration of geographical 262
indications and designations of origin (EC 1107/96). The two latter regulations supported and 263
acknowledged the quality of the hazelnut fruit, but other measures favoured the hazelnut expansion 264
too. A direct Community aid to farmers producing hazelnuts was granted since the 2003 (EC 265
1782/03) and more recently (2007) the hazelnut variety cultivated in the Langhe region was 266
registered within the Community Plant Variety Office (CPVO) with a new name (“Tonda gentile 267
Trilobata”) for a more efficient preservation. All these European regulations were locally 268
acknowledged by the PSR 2007-2013 (Action 214) rural development plan (Regione Piemonte, 269
2009). The success of hazelnut orchard in the Langhe region was favoured by the increasing 270
demand for quality products related to the sweet factory market (Garrone and Vacchetti, 1994; 271
Cova and Pace, 2006) that together with the EU food-labelling (PGI) policy facilitated the 272
“Nocciola Piemonte” to survive against stronger producer countries such as Turkey (Reis and 273
Yomralioglu, 2006). In fact, according to FAO database (FAOSTAT, 2010), hazelnut plantations in 274
Turkey increased their surface by 40% during the 1961 – 2000 time span. A smaller increase (23%) 275
of hazelnut surface in Italy and a stable situation in Spain have been observed in the same period. 276
277
4.4 Management issue
278
Hazelnut cultivation in the Langhe region expanded at the expenses of other orchards, fallow lands, 279
grasslands and generally in places where the cultivation of grapes was less remunerative (Valentini 280
and Me, 2002). Particularly important is the decline of crops in hilly regions such as Langhe that 281
resulted less productive than those located in flat areas (Comunità Montana Alta Langa 2009). The 282
rapid expansion of hazelnut plantations is radically transforming the rural landscape of Langhe 283
region and its further expansion could result in a use of not suitable areas for its cultivation. This 284
12
phenomenon has been recently observed in Turkey, where the hazelnut production exceeds the 285
demand mainly due to a lack in land use policy (Reis and Yomralioglu, 2006; Aydinoglu, 2010). 286
Medium-term consequence for this may include the abandonment of no more profitable hazelnut 287
orchards and consequent land degradation. These potential problems could be averted if the 288
effective quality of Langhe hazelnut production will continue to be achieved and adequately 289
protected. This paper highlights the potential of change detection investigations as a support for 290
national and international bodies in evaluating rural policies for valuable agriculture (London 291
Economics, 2008) and their effects on landscape, production and society (Westhoek et al., 2006; 292
Martinez-Casasnovas et al., 2010; Van Berkel and Verburg, 2011). Future researches should extend 293
the study area to give an accurate estimation of rural landscape dynamics filling the research gap 294
regarding remote detection of hazelnut cultivations expansion in the Mediterranean area (Franco, 295
1997; Reis and Yomralioglu, 2006). Moreover, socio-economic factors should be integrated in the 296
analysis of the drivers of land use change in order to achieve a more coherent and complete study of 297
the process under study (e.g. De Aranzabal et al., 2008; Tzanopoulos et al., in press). 298
299
Acknowledgments
300
The authors wish to thank the anonymous reviewers for their valuable comments and suggestions to 301
improve the manuscript and the Italian National Research Council (CNR-IRPI) for providing aerial 302
photographs and calibration certificates. Danilo Godone and Matteo Garbarino were funded by 303
Piemonte Region, DGR 57/OC of January 28, 2002. 304 305 306 307 308 309 310
13 311
312
Tables
313
Table 1 Comparison of root mean square errors obtained through the orthorectification of the 1954 314
aerial photographs with calibration certificate or with focal length only. 315 316 ResX (m) ResY (m) ResZ (m) |ResX| (m) |ResY | (m) |Res Z| (m) Calibration certificate Mean -0.11 -0.95 -1.29 2.45 1.80 1.51 RMS 3.12 2.07 1.48 1.65 1.25 1.22 Focal length Mean 1.36 4.27 14.86 8.42 16.73 18.78 RMS 13.51 19.95 15.86 10.10 9.63 9.88 317 318 319 320 321 322 323 324 325 326 327 328
14 329
330
Table 2 Description of the land use/land cover (LUCL) categories adopted in the aerial images 331
classification at Diano site. 332
333
LULC class Abbreviation Description
Forests FO Forest patches or single trees, excluding arboriculture plantations
Fields FI Cultivated or uncultivated grasslands and polygons not univocally identified Vineyards VI Surfaces cultivated with vine
Urban UR Human infrastructures and buildings Orchards OR Arboriculture and hazelnut orchards Waters WA Water bodies (lakes and rivers) 334 335 336 337 338 339 340 341 342 343 344 345 346 347
15 348
349
Table 3 Metrics on landscape level (McGarigal and Marks 1995) computed for the Diano study site 350
at two periods (1954-2000 land use maps). 351
352 353
Metrics (abbreviation) Component measured Units Land Use maps (2651.59 ha)
1954 2000 Patch Density (PD) Density n/100 ha 213.26 344.86
Patch Area Mean (AREA_MN) Area ha 0.47 0.29
Largest Patch Index (LPI) Area % 4.25 3.68
Edge Density (ED) Edge m/ha 486.95 411.43
Landscape Shape Index (LSI) Edge - 63.78 54.05
Shape Index Mean (SHAPE_MN) Shape - 1.99 1.52
Contiguity Index Mean (CONTIG_MN) Shape - 0.66 0.41 Euclidean N.N. Distance (ENN_MN) Isolation m 7.06 4.39 Contagion Index (CONTAG) Contagion % 60.08 53.55 Simpson's Diversity Index (SIDI) Diversity - 0.69 0.77 354 355 356 357 358 359 360
16
Table 4 Transition matrix showing land cover changes (ha) from 1954 to 2000. Values are 361
expressed in hectares and in percent (in parentheses) relative to the total area of the class in 1954. 362
363
Land uses in 2000
Land uses
in 1954
Forests Fields Vineyards Urban Orchards Water Total area (ha)
Forests 388.21 (60%) 58.77 (9%) 56.07 (9%) 41.93 (6%) 105.17 (16%) 0.89 (<1%) 651.04 (25%) Fields 160.81 (15%) 221.19 (20%) 268.69 (24%) 72.50 (7%) 375.24 (34%) 0.77 (<1%) 1099.20 (41%) Vineyards 78.69 (11%) 83.34 (11%) 370.93 (50%) 36.59 (5%) 165.09 (22%) 0.18 (<1%) 734.82 (28%) Urban 12.16 (11%) 19.03 (17%) 13.85 (12%) 54.62 (48%) 14.19 (12%) 0.16 (<1%) 114.00 (4%) Orchards 6.16 (18%) 1.25 (4%) 3.66 (10%) 1.56 (4%) 22.31 (64%) 0.00 (<1%) 34.92 (1%) Water 7.20 (41%) 2.97 (17%) 0.05 (<1%) 1.74 (10%) 4.16 (24%) 1.49 (8%) 17.60 (1%)
Total area (ha) 653.23 (25%) 386.54 (15%) 713.23 (27%) 208.93 (8%) 686.16 (26%) 3.49 (<1%) 2651.59 (100%)
364 365 366 367 368 369
Table 5 Total land surface area occupied by hazelnut orchards in Italy. Data are given as a whole 370
and divided by region (ISTAT, 2012). Statistical records are also reported for Turkey (Turkish 371 Statistical Institute, 2009). 372 Italian Regions 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 Data source Piemonte 8042 8042 8042 8043 9211 9519 9718 10531 11671 12366 12388 12142 12133 ISTAT
17 Campania 24805 24841 25417 25064 23496 22872 22831 22834 22828 22819 22849 12616 22787 ISTAT Lazio 18930 18878 18949 18999 19033 18996 18974 18985 18929 18914 18968 18969 19008 ISTAT Sicilia 15878 15730 15368 15368 15431 15146 15090 15080 16482 14930 16075 14350 14740 ISTAT Others 2140 2134 2058 2071 2094 2080 2244 2245 2394 2011 1749 1521 1814 ISTAT Italy 69795 69625 69834 69545 69265 68613 68857 69675 72304 71040 72029 59598 70482 ISTAT Turkey 530700 549500 550000 560000 600000 621200 621200 621200 621200 621200 - - - TSI 373 374 375 376 377 378 379 380 Figure captions 381
Figure 1 Location of the 2651 ha study area (white dot) within Piedmont region (upper image) and 382
its topography (lower image) with main rivers and settlements. 383
Figure 2 Land use classifications of 1954 (upper map) and 2000 (lower map) orthoimages. 384
Figure 3 Land uses expressed as proportion of the total study site surface in 1954 (black bars) and 385
2000 (grey bars) at Diano site. 386
Figure 4 Agricultural statistics of Cuneo (Piedmont, Italy) province in the 1971-2001 period. 387
18 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 References 407 408
Acosta, A., Carranza, M.L., Giancola, M., 2005. Landscape Change and Ecosystem Classification 409
in a Municipal District of a Small City (Isernia, Central Italy). Environmental Monitoring and 410
Assessment 108, 323 - 335. 411
Aydinoglu, A.C., 2010. Examining environmental condition on the growth areas of Turkish 412
hazelnut (Corylus colurna L.). African Journal of Biotechnology 9, 6492-6502. 413
19
Amsden, A.H., 1991. Diffusion of Development: The Late-Industrializing Model and Greater East 414
Asia. The American Economic Review 81, 282 - 286. 415
Antrop, M., 2004. Landscape change and the urbanization process in Europe. Landscape and Urban 416
Planning 67, 9 – 26. 417
ARPA, 2012. Geologia e processi di dissesto. 418
<http://webgis.arpa.piemonte.it/elenco_servizi/index.html> 419
Becattini, G., Coltorti, F., 2006. Areas of large enterprise and industrial districts in the development 420
of post-war Italy: A preliminary survey. European Planning Studies 14, 1105 -1138. 421
Beni Culturali, 2011. National Aerial Photographic Archive - The Collection. 422
<http://www.iccd.beniculturali.it/index.php?en/165/le-collezioni> 423
Boccacci, P., Botta, R., 2009. Investigating the origin of hazelnut (Corylus avellana L.) cultivars 424
using chloroplast microsatellites. Genetic Resources and Crop Evolution 56, 851 - 859. 425
Bonet, A., 2004. Secondary succession of semi-arid Mediterranean old-fields in south-eastern 426
Spain: Insights for conservation and restoration of degraded lands. Journal of Arid Environments 427
56, 213 - 233. 428
Burgi, M., Russell, E.W.B., 2001. Integrative methods to study landscape changes. Land Use Policy 429
18, 9-16. 430
Cain, D., Riitters, K., Orvis, K., 1997. A multi-scale analysis of landscape statistics. Landscape 431
Ecology 12, 199 - 212. 432
Campana, S., Francovich, R., 2003. Landscape archaeology in Tuscany: Cultural Resource 433
Management, remotely sensed techniques, GIS based data integration and interpretation, in: Forte, 434
M., Williams, P. R. (Ed.), The Reconstruction of Archaeological Landscapes through Digital 435
Technologies. Archaeopress, Oxford, pp. 15 - 28. 436
Ceraudo, G., 2005. 105 Years of Archaeological Aerial Photography in Italy (1899 - 2004), in: 437
Bourgeois, J., Meganck, M. (Ed.), Aerial Photography and Archaeology 2003. A Century of 438
Information. Academia Press, Ghent University pp. 73 - 86. 439
20
Chandrasekhar, T., 1999. Change Detection Extension for ArcView. ESRI, Redlands. 440
Chiabrando, R., Fabrizio, E., Garnero, G., 2009. The territorial and landscape impacts of 441
photovoltaic systems: Definition of impacts and assessment of the glare risk. Renewable and 442
Sustainable Energy Reviews 13, 2441-2451. 443
Chiabrando, R., Fabrizio, E., Garnero, G., 2011. On the applicability of the visual impact 444
assessment OAISPP tool to photovoltaic plants. Renewable and Sustainable Energy Reviews 15, 445
845-850. 446
Cocca, G., Sturaro, E., Gallo, L., Ramanzin, M., 2012. Is the abandonment of traditional livestock 447
farming systems the main driver of mountain landscape change in Alpine areas? Land Use Policy 448
29, 878 - 886. 449
Comunità Montana Alta Langa, 2009. La coltivazione del nocciolo in Alta Langa. Linee guida per 450
una corilicoltura sostenibile. Comunità Montana Alta Langa, Bra, p. 66. 451
Conti, G., Fagarazzi, L., 2004. Sustainable mountain development and the key issue of 452
abandonment of marginal rural areas. Planum - The Journal of Urbanism 11, 1 - 20. 453
Corti, G., Cavallo, E., Cocco, S., Biddoccu, M., Brecciaroli, G., Agnelli, A., 2011. Evaluation of 454
Erosion Intensity and Some of Its Consequences in Vineyards from Two Hilly Environments Under 455
a Mediterranean Type of Climate, Italy, in: Godone, D., Stanchi, S. (Ed.), Soil Erosion Issues in 456
Agriculture. InTech, Rijeka, pp. 113 – 160. 457
Cots-Folch, R., Martinez-Casasnovas, J.A., Ramos, M.C., 2006. Land terracing for new vineyard 458
plantations in the northeastern Spanish Mediterranean region: Landscape effects of the EU council 459
regulation policy for vineyards’ restructuring. Agriculture Ecosystems and Environment 115, 88– 460
96. 461
Cova, B., Pace, S., 2006. Brand community of convenience products: new forms of customer 462
empowerment – the case “my Nutella The Community”. European Journal of Marketing 40, 1087 - 463
1105. 464
21
Cullotta, S., Barbera, G., 2011. Mapping traditional cultural landscapes in the Mediterranean area 465
using a combined multidisciplinary approach: Method and application to Mount Etna (Sicily; Italy). 466
Landscape and Urban Planning 100, 98-108. 467
Cushman, S.A., McGarigal, K., Neel, M.C., 2008. Parsimony in landscape metrics: Strength, 468
universality, and consistency. Ecological Indicators 8, 691-703. 469
De Aranzabal, I., Schmitz, M.F., Aguilera, P., Pineda, F.D., 2008. Modelling of landscape changes 470
derived from the dynamics of socio-ecological systems: A case of study in a semiarid 471
Mediterranean landscape. Ecological Indicators 8, 672-685. 472
Delmastro, M., 2005. An investigation into the quality of wine: Evidence from piedmont. Journal of 473
Wine Research 16, 1-17. 474
European Commission, 2004. The Common Agricultural Policy explained. European Commission 475
Directorate General for Agriculture, Brussels. 476
Falcucci, A., Maiorano, L., Boitani, L., 2007. Changes in land-use/land-cover patterns in Italy and 477
their implications for biodiversity conservation. Landscape Ecology 22, 617 – 631. 478
Fauri, F., 1996. The Role of Fiat in the Development of the Italian Car Industry in the 1950's. The 479
Business History Review 70, 167 - 206. 480
Fairbanks, D.H.K., Hughes, C.J., Turpie, J.K., 2004. Potential impact of viticulture expansion on 481
habitat types in the Cape Floristic Region, South Africa. Biodiversity and Conservation 13, 1075– 482
1100. 483
FAOSTAT, 2010. <http://faostat.fao.org/site/567/DesktopDefault.aspx?PageID=567#ancor> 484
Franco, S., 1997. Use of remote sensing to evaluate the spatial distribution of hazelnut cultivation: 485
Results of a study performed in an Italian production area. Acta Horticulturae 445, 381 - 388. 486
Garrone, W., Vacchetti, M., 1994. Hazelnut quality in relation with the requirements of 487
confectionary industry, in: Me, G., Radicati, L. (Ed.), III International Congress on Hazelnut, 488
Wageningen pp. 641 - 647. 489
22
Gennaretti, F., Ripa, M.N., Gobattoni, F., Boccia, L., Pelorosso, R., 2011. A methodology proposal 490
for land cover change analysis using historical aerial photos. Journal of Geography and Regional 491
Planning 4, 542 - 556. 492
Hatton, T.J., Williamson, J.G., 1994. What drove the mass migrations from Europe in the late 493
nineteenth century? Population and Development Review 20, 1 - 27. 494
IRPI, 2012. Aerial photo archive. <http://www.irpi.to.cnr.it/index_en.php?page=aerofototeca_en.> 495
Kaye, C.A., 1957. Military Geology in the United States Sector of the European Theater of 496
Operations During World War II. Geological Society of America Bulletin 68, 47 - 54. 497
Landis, J.R., Koch, G.G. 1977. The measurement of observer agreement for categorical data. 498
Biometrics 33, 159-174.London Economics, 2008. Evaluation of the CAP policy on protected 499
designations of origin (PDO) and protected geographical indications (PGI). European Commission - 500
Agriculture and Rural Development, p. 275. 501
Martinez-Casasnovas, J.A., Ramos, M.C., Cots-Folch, R., 2010. Influence of the EU CAP on terrain 502
morphology and vineyard cultivation in the Priorat region of NE Spain. Land Use Policy 27, 11 - 503
21. 504
Marull, J., Pinob, J., Tello, E., Cordobilla, M.J., 2010. Social metabolism, landscape change and 505
land-use planning in the Barcelona Metropolitan Region. Land Use Policy 27, 497 – 510. 506
McClure, J.T., Griffiths, G.H., 2002. Historic Landscape Reconstruction and Visualisation, West 507
Oxfordshire, England. Transactions in GIS 18, 289 - 352. 508
McGarigal, K., Marks, B.J., 1995. FRAGSTATS: spatial pattern analysis program for quantifying 509
landscape structure, Portland. 510
Meeus, J.H.A., Wijermans, M.P., Vroom, M.J., 1990. Agricultural landscapes in Europe and their 511
transformation. Landscape Urban Planning 18, 289 - 352. 512
Merenlender, A.M. 2000. Mapping vineyard expansion provides information on agriculture and the 513
environment. California Agriculture 54, 7–12. 514
23
Mikhail, M.E., Bethel, J.S., Mc Glone, J.C., 2001. Introduction to Modern Photogrammetry. John 515
Wiley and Sons, New York, p. 479. 516
Morgan, J.L., Gergel, S.E., Coops, N.C., 2010. Aerial photography: a rapidly evolving tool for 517
ecological management. BioScience 60, 47 - 59. 518
Neel, M., McGarigal, K., Cushman, S., 2004. Behavior of class-level landscape metrics across 519
gradients of class aggregation and area. Landscape Ecology 19, 435 - 455. 520
Peel, M.C., Finlayson, B.L., McMahon, T.A., 2007. Updated world map of the Köppen-Geiger 521
climate classification. Hydrology and Earth System Sciences 11, 1633 - 1644. 522
Peroni, P., Ferri, F., Avena, G.C., 2000. Temporal and spatial changes in a mountainous area of 523
central Italy. Journal of Vegetation Science 11, 505 - 514. 524
Pinto-Correia, T., 2000. Future development in Portuguese rural areas: how to manage agricultural 525
support for landscape conservation? Landscape and Urban Planning 50, 95 - 106. 526
Pinto-Correia, T., 1993. Land abandonment: changes in the land use patterns around the 527
Mediterranean Basin. International Centre for Advanced Mediterranean Agronomic Studies 528
(CIHEAM) Zaragoza, pp. 97 – 112. 529
Rabbinge, R., Van Latesteijn, H.C., 1992. Long-term options for land use in the European 530
community. Agricultural Systems 40, 195 - 210. 531
Reed, H.P., 1946. The Development of the Terrain Model in the War. Geographical Review 36, 632 532
- 652. 533
Regione Piemonte, 2009. Programma di sviluppo rurale PSR 2007-2013. Regione Piemonte, 534
Torino, p. 656. 535
Reis, S., Yomralioglu, T., 2006. Detection of current and potential hazelnut plantation areas in 536
Trabzon, North East Turkey using GIS and RS. Journal of Environmental Biology 27, 653 - 659. 537
Riitters, K., O’Neill, R., Hunsaker, C., Wickham, J., Yankee, D., Timmins, S., Jones, K., Jackson, 538
B., 1995. A factor analysis of landscape pattern and structure metrics. Landscape Ecology 10, 23 - 539
39. 540
24
Rocchini, D., Metz, M., Frigeri, M., Delucchi, L., Marcantonio, M., Neteler, M., 2012. Robust 541
rectification of aerial photographs in an open source environment. Computer & Geosciences 39, 542
145 - 151. 543
Scardozzi, G., 2008. Old high resolution satellite images for landscape archaeology: case studies 544
from Turkey and Iraq, in: Michel, U., Civco, D. L., Ehlers, M., Kaufmann, H. J. (Ed.), Remote 545
Sensing for Environmental Monitoring, GIS Applications, and Geology VIII. 546
Schiefer, E., Gilbert, R., 2007. Reconstructing morphometric change in a proglacial landscape using 547
historical aerial photography and automated DEM generation. Geomorphology 88, 167 - 178. 548
Sitzia, T., Semenzato, P., Trentanovi, G., 2010. Natural reforestation is changing spatial patterns of 549
rural mountain and hill landscapes: A global overview. Forest Ecology and Management 259, 1354 550
- 1362. 551
Sluiter, R., de Jong, S.M., 2007. Spatial patterns of Mediterranean land abandonment and related 552
land cover transitions. Landscape Ecology 22, 559 - 576. 553
Stoate, C., Boatman, N.D., Borralho, R.J., Carvalho, C.R., Snoo, G.R.d., Eden, P., 2001. Ecological 554
impacts of arable intensification in Europe. Journal of Environmental Management 63, 337-365. 555
Tanrivermis, H., 2003. Agricultural land use change and sustainable use of land resources in the 556
Mediterranean region of Turkey. Journal of Arid Environments 54, 553 - 564. 557
Tischendorf, L., 2001. Can landscape indices predict ecological processes consistently? Landscape 558
Ecology 16, 235 - 254. 559
Trotter, A., 1921. Contributo alla storia colturale del nocciuolo nella Campania, Congresso di 560
Arboricoltura Meridionale, Napoli (Italy), pp. 3 - 19. 561
Turner, M.G., Gardner, R.H., 1990. Quantitative Methods in Landscape Ecology. Springer, New 562
York. 563
Tzanopoulos, J., Jones, P.J., Mortimer, S.R., in press. The implications of the 2003 Common 564
Agricultural Policy reforms for land-use and landscape quality in England. Landscape and Urban 565
Planning. 566
25
UNESCO, 2012. Wine Grape landscapes: Langhe, Roero, Monferrato and Valtellina. 567
<http://whc.unesco.org/en/tentativelists/5001/> 568
Valentini, N., Me, G., 2002. Attualità e problematiche della coltura del nocciolo in Italia: la 569
situazione piemontese, Atti del 2° Convegno Nazionale sul Nocciolo, Giffoni Valle Piana (SA), pp. 570
133 - 140. 571
Van Berkel, D.B., Verburg, P.H., 2011. Sensitising rural policy: Assessing spatial variation in rural 572
development options for Europe. Land Use Policy 28, 447 - 459. 573
Van Rompaey, A.J.J., Govers, G., Van Hecke, E., Jacobs, K., 2001. The impacts of land use policy 574
on the soil erosion risk: a case study in central Belgium. Agriculture, Ecosystems and Environment 575
83, 83 – 94. 576
Vos, W., Meekes, H., 1999. Trends in European cultural landscape development: perspectives for a 577
sustainable future. Landscape Urban Planning 46, 3 – 14. 578
Vuorela, N., Alho, P., Kalliola, R., 2002. Systematic Assessment of Maps as Source Information in 579
Landscape-change Research. Landscape Research 27, 141 - 166. 580
Westhoek, H.J., van den Berg, M., Bakkes, J. A., 2006. Scenario development to explore the future 581
of Europe's rural areas. Agriculture, Ecosystems & Environment 114, 7 - 20. 582
Zomeni, M., Tzanopoulos, J., Pantis, J. D., 2008. Historical analysis of landscape change using 583
remote sensing techniques: An explanatory tool for agricultural transformation in Greek rural areas. 584
Landscape and Urban Planning 86, 38- 46. 585
